This Competitive Renewal will build upon biomarker discoveries from the previous period. In four Aims we will focus upon the overall hypothesis that inflammatory candidate biomarkers identify patients at risk for incident symptomatic knee OA (SKOA), or those patients with established SKOA, who are at increased risk for disease progression. Together with our collaborators on this grant, we propose to validate and replicate our initial discoveries using four OA cohorts: 1) the NIH Osteoarthritis Initiative (OAI, M. Hochberg); 2) a second cycle OAI+expanded (e)NYU (OAI+eNYU) discovery cohort using high quality radiographs; 3) the Johnston County Osteoarthritis Project (JoCo, J. Jordan); and 4) the Genetics of Generalized Osteoarthritis longitudinal study (GOGO Long, V. Kraus). In Aim 1, we will determine whether candidate Genomic biomarkers (peripheral blood leukocyte expression of inflammatory genes) or plasma Protein/Lipid (GPL) biomarkers predict increased risk of disease severity or progression in subjects with symptomatic radiographic knee OA. We will: a) validate discovered GPL biomarker panel from NYU in the OAI; b) construct a panel of GPL biomarkers from OAI+eNYU; c) replicate this panel in GOGO Long and JoCo. In Aim 2 we will determine whether candidate genetic biomarkers (SNPs from GWAS assays that map to inflammatory genes and exonic variants from whole-exome sequencing assay) predict increased risk of disease progression in subjects with radiographic knee OA. We will: a) construct a panel of SNP biomarkers from OAI+eNYU; b) validate this panel in GOGO+JoCo; c) construct a panel of exonic variant biomarkers from eNYU+OAI; d) validate this panel by holdout in eNYU+OAI. In Aim 3 we will perform an integrative analysis, applying a predictive multi-assay analyses approach to identify potential pathogenic interactions among genetic variations, gene expression and plasma protein/lipid expression. We will: a) perform integrative modeling using GPL+SNP biomarkers in OAI+eNYU; b) validate this model in GOGO and JoCo; c) perform integrative modeling using GPL+SNP+exonic variant biomarkers in OAI+eNYU. In Aim 4 we will determine whether genomic, genetic and protein/lipid biomarkers predict increased risk of incident symptomatic knee OA. Relevance: Osteoarthritis is a common disease that affects multiple joints, resulting in joint destruction, loss of function and disability. Base on our preliminary data we expect that this proposal will identify prognostic biomarker test(s) that can predict individuals at risk for disease onset or of progressing rapidly to joint destruction. Validation of such prognostic biomarkers will facilitate the development new disease-modifying OA drugs, and, in the future, could enhance clinical management of the disease, as in decisions regarding administration of drugs.